54 BULLETIN 1090, U. S. DEPARTMENT OF AGRICULTURE. 



percentage of young which reached weaning. Family 3 agrees with 

 Families 9 and 11 in combining rapid growth with inability to raise the 

 young successfully. It was doubtless this poor success in raising the 

 young which caused Families 3 and 11 to be among the five families 

 which were extinct at the end of 1915. 



Family 2, during the second period, had a combination of char- 

 acters the opposite of that in Families 9 and 1 1 . With the smallest 

 weight and small litters it combined the greatest regularity in pro- 

 ducing litters and the best record in raising the young which were 

 born alive. Family 17 was consistently of this type in most respects. 

 During both periods it produced small litters and small pigs, but 

 pigs which were easily raised. This combination seemed to be more 

 fortunate than that of Families 9, 11, and 3, since Family 2 became 

 the most numerous of all the families after 1915, and Family 17 was a 

 large family, while, as noted above, Families 3 and 11 were among 

 the first to become extinct and Family 9 was always a small family. 



It was shown earlier that the only significant correlations between 

 the different groups of characters were those between birth weight or 

 gain and size of litter ( + 0.26 and + 0.37 in the first period, both -f 0.62 

 in the second) . The impossibility of considering these correlations as 

 an indication of heredity of general vigor may be seen by comparing 

 Families 9, 11, and 3 with Families 2 and 17. As we have just seen, 

 the latter families appeared to be the more successful, in spite of 

 their great inferiority in size of litter and weight. 



There is a possible cause of correlation between characters which 

 should be mentioned. It will be remembered that the experi- 

 ment started with 23 different females, but only 9 males. If one of 

 the male ancestors of several families happened to transmit two of 

 the characters in an extreme degree, it would tend to bring about a 

 correlation in these characters among the families, which would have 

 no significance as an indication of inheritance of general vigor or of a 

 common physiological factor. It may not be a coincidence that the 

 three families which traced back in all lines (Families 9 and 11) or 

 in the principal line (Family 13) to a certain foundation male ancestor 

 (Male 3) should be the three leading families in both size of litter and 

 weight. Again, two of the four families descended from Male 1 

 (Families 1 and 2 in the group 1, 2, 3, and 7) are characterized by 

 remarkably small litters and light weight. If we suppose that 

 Male 3 transmitted both large size of litter and great weight in a 

 remarkable degree, and perhaps that the converse was true of Male 1, 

 we can account for the correlation between these characters observed 

 in the present data. 



The other groups of families with common male ancestors, namely, 

 Families 17 to 24, 31 and 32, 35 and 36, were not clearly differen- 

 tiated, but it may well have been that Males 3, 9, and 11 were medioc- 



